University of Twente Student Theses
Transparency and Efficiency in Credit Risk Assessment of Alternative Financing : A Green AI Approach
Radzkova, Katsiaryna (2024) Transparency and Efficiency in Credit Risk Assessment of Alternative Financing : A Green AI Approach.
|
PDF
524kB |
| Abstract: | Peer-to-Peer (P2P) lending platforms utilize Machine Learning (ML) to predict loan defaults effectively. Nevertheless, the implementation and training of these models require significant computational resources, which raises environmental issues. In addition to analyzing the application of ML models for default risk prediction, the research proposes ways to minimize the environmental impact by following Green AI principles. The goal of the study is to discover algorithms that maintain a balance between predicted accuracy and sustainability by examining four ML models: XGBoost, Random Forests, Decision Trees, and Logistic Regression. Additionally, the research explores strategies of applying Explainable AI techniques to increase transparency of the models. The results of the research demonstrate that although Random Forest model have a high default prediction accuracy, its shows large environmental impact in terms of CO2 emissions. However, despite lower performance, Logistic Regression offers a reasonable balance between accuracy and sustainability of the model. |
| Item Type: | Essay (Bachelor) |
| Faculty: | EEMCS: Electrical Engineering, Mathematics and Computer Science |
| Subject: | 54 computer science, 83 economics |
| Programme: | Computer Science BSc (56964) |
| Link to this item: | https://purl.utwente.nl/essays/100947 |
| Export this item as: | BibTeX EndNote HTML Citation Reference Manager |
Show download statistics for this publication
Show download statistics for this publicationRepository Staff Only: item control page